Autonomous Vehicle Control Profiling With Standardized Driving Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional autonomous vehicle control systems lack a systematic and comprehensive method for quantitative analysis, leading to time-consuming and incomplete assessments, and there is a need for uniform and automatic test data profiling tools to improve motion planning and control.
Innovation Solution
A computer-implemented method that extracts and filters driving records from autonomous vehicles, generates standardized metrics, and updates controllers based on these metrics, using a data processing system to align and analyze records for improved performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual analysis methods are used for autonomous vehicle control system, then analysis can be performed, but the process is time-consuming and incomplete
Solution Approach 1:
The patent replaces manual mechanical analysis processes with an automated computer-implemented system that extracts driving records, generates standardized metrics, and updates controllers automatically. This substitution eliminates the time-consuming manual selection of features and non-uniform analysis methods, directly resolving the contradiction between analysis completeness and time consumption.
Solution Approach 2:
The system enables self-service analysis by automatically processing driving records and generating performance metrics without requiring manual intervention. The automated extraction and filtering of driving records, along with standardized metric generation, allows the control system to improve itself through systematic self-analysis, significantly increasing productivity while reducing time loss.
2Ease of manufacture
If same motion planning and control is applied to all vehicle types, then implementation is simplified, but accuracy and smoothness deteriorate
Solution Approach 1:
The patent applies local quality by extracting and analyzing driving records specific to each ADV type and controller type combination. The system generates standardized metrics tailored to each vehicle-controller pair, enabling customized control improvements that account for specific vehicle characteristics while maintaining a uniform analytical framework. This resolves the contradiction by providing type-specific accuracy within a standardized process.
Solution Approach 2:
The system changes parameters by generating type-specific standardized metrics from filtered driving records for each ADV type and controller type. Rather than applying identical control parameters universally, the system adapts control strategies based on extracted performance characteristics from specific vehicle types, thereby improving motion planning accuracy while maintaining implementation feasibility through systematic parameter adaptation.
3Reliability
If comprehensive test data profiling is performed for each functional module, then analysis completeness improves, but system complexity increases
Solution Approach 1:
The patent applies universality by creating a standardized metric generation system that handles multiple functional modules (perception, localization, planning, control) through a unified approach. The same computer-implemented method extracts driving records, filters data, and generates standardized metrics across all modules, reducing system complexity while maintaining comprehensive analysis completeness through multi-functional capability.
4Adaptability or versatility
If non-uniform analysis methods are used, then flexibility is maintained, but uniformity and consistency across iterations are lost
Solution Approach 1:
The patent resolves this contradiction by standardizing analysis parameters through automated metric generation. The system extracts driving records and applies uniform filtering and metric generation procedures across all analyses, ensuring consistency and repeatability. Standardized metrics provide uniformity across iterations while the system remains adaptable to different ADV types and controller types through parameterized analysis configurations.
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
Systems and methods are disclosed for collecting driving data from simulated autonomous driving vehicle (ADV) driving sessions and real-world ADV driving sessions. The driving data is processed to exclude manual (human) driving data and to exclude data corresponding to the ADV being stationary (not driving). Data can further be filtered based on driving direction: forward or reverse driving. Driving data records are time stamped. The driving data can be aligned according to the timestamp, and then a standardized set of metrics is generated from the collected, filtered, and time-aligned data. The standardized set of metrics are used to grade the performance the control system of the ADV, and to generate an updated ADV controller, based on the standardized set of metrics. The methods provide a systematic, comprehensive and automatic test tool for autonomous vehicle control system.